The present study aimed to explore the predictive ability of frailty index in mortality and need for care among older Swedes. Frailty index was also compared with multimorbidity and limitations in ADLs which are the traditional benchmarks for the mortality and need for care prediction. The frailty index showed an association with mortality among females with the shorter, but not the longer follow-up time (≤17 and >17 years, respectively). This difference may originate from the fact that the females in the former group were also older (median age 74 vs. 56 years). Thus, this difference may also reflect the fact that the frailty index predicts mortality only among old females, but not among the young in our population. However, due to the choice of …show more content…
One limitation of our study is the construction of the frailty index. It was based on self-reported data and some of the questionnaires may have been answered by a third person, introducing less reliability into the measure. Similarly, some health conditions such as severe cognitive decline, may limit participation in this study. Due to this factor, the prevalence of frailty and its relationship with mortality and need for care may have been under-estimated in our study. In addition, not all potential confounders or mediating factors could be tested, such as economic status, social networks, dietary habits, and interpersonal relationships. Furthermore, the number of items and the content vary from one study to another due to the lack of internationally established questionnaires to make the approach standardized. Although all 42 health deficits that were involved in the frailty index construction fulfilled the Rockwood criteria (13), the results are still not easily comparable with other similar studies. As the study was based on self-reported questionnaire items, we did not consider biomedical indicators as frailty variables, and this could also be a